Graph Theory-Based Brain Network Connectivity Analysis and Classification of Alzheimer’s Disease

Brain Networks ◽

Network Connectivity ◽

Brain Regions ◽

Brain Network ◽

Fiber Tracts ◽

Alzheimer’s Disease (AD) is basically a progressive neurodegenerative disorder associated with abnormal brain networks that affect millions of elderly people and degrades their quality of life. The abnormalities in brain networks are due to the disruption of White Matter (WM) fiber tracts that connect the brain regions. Diffusion-Weighted Imaging (DWI) captures the brain’s WM integrity. Here, the correlation betwixt the WM degeneration and also AD is investigated by utilizing graph theory as well as Machine Learning (ML) algorithms. By using the DW image obtained from Alzheimer’s Disease Neuroimaging Initiative (ADNI) database, the brain graph of each subject is constructed. The features extracted from the brain graph form the basis to differentiate between Mild Cognitive Impairment (MCI), Control Normal (CN) and AD subjects. Performance evaluation is done using binary and multiclass classification algorithms and obtained an accuracy that outperforms the current top-notch DWI-based studies.

Brain Network Modeling Based on Mutual Information and Graph Theory for Predicting the Connection Mechanism in the Progression of Alzheimer’s Disease

Entropy ◽

10.3390/e21030300 ◽

2019 ◽

Vol 21 (3) ◽

pp. 300 ◽

Cited By ~ 4

Author(s):

Shuaizong Si ◽

Bin Wang ◽

Xiao Liu ◽

Chong Yu ◽

Chao Ding ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Graph Theory ◽

Mutual Information ◽

Brain Networks ◽

Brain Network ◽

Network Efficiency ◽

Characteristic Path Length ◽

Anatomical Space ◽

Alzheimer’s disease (AD) is a progressive disease that causes problems of cognitive and memory functions decline. Patients with AD usually lose their ability to manage their daily life. Exploring the progression of the brain from normal controls (NC) to AD is an essential part of human research. Although connection changes have been found in the progression, the connection mechanism that drives these changes remains incompletely understood. The purpose of this study is to explore the connection changes in brain networks in the process from NC to AD, and uncovers the underlying connection mechanism that shapes the topologies of AD brain networks. In particular, we propose a mutual information brain network model (MINM) from the perspective of graph theory to achieve our aim. MINM concerns the question of estimating the connection probability between two cortical regions with the consideration of both the mutual information of their observed network topologies and their Euclidean distance in anatomical space. In addition, MINM considers establishing and deleting connections, simultaneously, during the networks modeling from the stage of NC to AD. Experiments show that MINM is sufficient to capture an impressive range of topological properties of real brain networks such as characteristic path length, network efficiency, and transitivity, and it also provides an excellent fit to the real brain networks in degree distribution compared to experiential models. Thus, we anticipate that MINM may explain the connection mechanism for the formation of the brain network organization in AD patients.

Alzheimer’s disease under the purview of Graph Theory Centric Genetic Networks

BRAIN. BROAD RESEARCH IN ARTIFICIAL INTELLIGENCE AND NEUROSCIENCE ◽

10.18662/brain/12.2/199 ◽

2021 ◽

Vol 12 (2) ◽

Author(s):

Yegnanarayanan Venkatraman ◽

◽

Narayanaa Y Krithicaa ◽

Valentina E. Balas ◽

Marius M. Balas ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Graph Theory ◽

Network Theory ◽

Memory Loss ◽

Amyloid Plaque ◽

Brain Regions ◽

Protein Protein Interaction ◽

Communication Demands ◽

Notice that the synapsis of brain is a form of communication. As communication demands connectivity, it is not a surprise that "graph theory" is a fastest growing area of research in the life sciences. It attempts to explain the connections and communication between networks of neurons. Alzheimer’s disease (AD) progression in brain is due to a deposition and development of amyloid plaque and the loss of communication between nerve cells. Graph/network theory can provide incredible insights into the incorrect wiring leading to memory loss in a progressive manner. Network in AD is slanted towards investigating the intricate patterns of interconnections found in the pathogenesis of brain. Here, we see how the notions of graph/network theory can be prudently exploited to comprehend the Alzheimer’s disease. We begin with introducing concepts of graph/network theory as a model for specific genetic hubs of the brain regions and cellular signalling. We begin with a brief introduction of prevalence and causes of AD followed by outlining its genetic and signalling pathogenesis. We then present some of the network-applied outcome in assessing the disease-signalling interactions, signal transduction of protein-protein interaction, disturbed genetics and signalling pathways as compelling targets of pathogenesis of the disease.

Comparison of machine learning approaches for enhancing Alzheimer’s disease classification

PeerJ ◽

10.7717/peerj.10549 ◽

2021 ◽

Vol 9 ◽

pp. e10549

Author(s):

Qi Li ◽

Mary Qu Yang

Keyword(s):

Machine Learning ◽

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Structural Changes ◽

Spatial Information ◽

Brain Regions ◽

Disease Classification ◽

Support Vector ◽

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder, accounting for nearly 60% of all dementia cases. The occurrence of the disease has been increasing rapidly in recent years. Presently about 46.8 million individuals suffer from AD worldwide. The current absence of effective treatment to reverse or stop AD progression highlights the importance of disease prevention and early diagnosis. Brain structural Magnetic Resonance Imaging (MRI) has been widely used for AD detection as it can display morphometric differences and cerebral structural changes. In this study, we built three machine learning-based MRI data classifiers to predict AD and infer the brain regions that contribute to disease development and progression. We then systematically compared the three distinct classifiers, which were constructed based on Support Vector Machine (SVM), 3D Very Deep Convolutional Network (VGGNet) and 3D Deep Residual Network (ResNet), respectively. To improve the performance of the deep learning classifiers, we applied a transfer learning strategy. The weights of a pre-trained model were transferred and adopted as the initial weights of our models. Transferring the learned features significantly reduced training time and increased network efficiency. The classification accuracy for AD subjects from elderly control subjects was 90%, 95%, and 95% for the SVM, VGGNet and ResNet classifiers, respectively. Gradient-weighted Class Activation Mapping (Grad-CAM) was employed to show discriminative regions that contributed most to the AD classification by utilizing the learned spatial information of the 3D-VGGNet and 3D-ResNet models. The resulted maps consistently highlighted several disease-associated brain regions, particularly the cerebellum which is a relatively neglected brain region in the present AD study. Overall, our comparisons suggested that the ResNet model provided the best classification performance as well as more accurate localization of disease-associated regions in the brain compared to the other two approaches.

3265 Analysis of High-Dimensional Patient Data in Characterizing Alzheimer’s Disease Progression

Journal of Clinical and Translational Science ◽

10.1017/cts.2019.10 ◽

2019 ◽

Vol 3 (s1) ◽

pp. 3-3

Author(s):

Daniel Baer ◽

Andrew B. Lawson ◽

Brandon Vaughan ◽

Jane E. Joseph

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Graph Theory ◽

Cognitive Decline ◽

Brain Regions ◽

Clustering Coefficient ◽

Eigenvector Centrality ◽

Research Hypothesis ◽

The Brain ◽

Over Time

OBJECTIVES/SPECIFIC AIMS: Our research hypothesis is that resting state fMRI (rsfMRI) data can be used to identify regions of the brain which are associated with cognitive decline in patients – thereby providing a tool by which to characterize AD progression in patients. METHODS/STUDY POPULATION: We used data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) to analyze Mini-Mental State Examination (MMSE) questionnaire scores from 14 patients diagnosed with AD at two measurement occasions. RsfMRI data was available at the first of these occasions for these patients. These rsfMRI data were summarized into 264 node-based graph theory measures of clustering coefficient and eigenvector centrality. To address our research hypothesis, we modeled changes in patient MMSE scores over time as a function of these rsfMRI data, controlling for relevant confounding factors. This model accounted for the high-dimensionality of our predictor data, the longitudinal nature of the outcome, and our desire to identify a subset of regions in the brain most associated with the MMSE outcome. RESULTS/ANTICIPATED RESULTS: The use of either the clustering coefficient or eigenvector centrality rsfMRI predictors in modeling MMSE scores for patients over time resulted in the identification of different subsets of brain regions associated with cognitive decline. This suggests that these predictors capture different information on patient propensity for cognitive decline. Further work is warranted to validate these results on a larger sample of ADNI patients. DISCUSSION/SIGNIFICANCE OF IMPACT: We conclude that different rsfMRI graph theory measures capture different aspects of cognitive function and decline in patients, which could be a future consideration in clinical practice.

Brain Network Alterations in Alzheimer’s Disease Identified by Early-Phase PIB-PET

Contrast Media & Molecular Imaging ◽

10.1155/2018/6830105 ◽

2018 ◽

Vol 2018 ◽

pp. 1-10 ◽

Cited By ~ 3

Author(s):

Liping Fu ◽

Linwen Liu ◽

Jinming Zhang ◽

Baixuan Xu ◽

Yong Fan ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Early Phase ◽

Brain Networks ◽

Normal Subjects ◽

Brain Network ◽

Highly Correlated ◽

Fdg Pet Ct ◽

Cognitively Normal Subjects ◽

The aim of this study was to identify the brain networks from early-phase 11C-PIB (perfusion PIB, pPIB) data and to compare the brain networks of patients with differentiating Alzheimer’s disease (AD) with cognitively normal subjects (CN) and of mild cognitively impaired patients (MCI) with CN. Forty participants (14 CN, 12 MCI, and 14 AD) underwent 11C-PIB and 18F-FDG PET/CT scans. Parallel independent component analysis (pICA) was used to identify correlated brain networks from the 11C-pPIB and 18F-FDG data, and a two-sample t-test was used to evaluate group differences in the corrected brain networks between AD and CN, and between MCI and CN. Our study identified a brain network of perfusion (early-phase 11C-PIB) that highly correlated with a glucose metabolism (18F-FDG) brain network and colocalized with the default mode network (DMN) in an AD-specific neurodegenerative cohort. Particularly, decreased 18F-FDG uptake correlated with a decreased regional cerebral blood flow in the frontal, parietal, and temporal regions of the DMN. The group comparisons revealed similar spatial patterns of the brain networks derived from the 11C-pPIB and 18F-FDG data. Our findings indicate that 11C-pPIB derived from the early-phase 11C-PIB could provide complementary information for 18F-FDG examination in AD.

Communicability distance reveals hidden patterns of Alzheimer’s disease

Network Neuroscience ◽

10.1162/netn_a_00143 ◽

2020 ◽

Vol 4 (4) ◽

pp. 1007-1029 ◽

Cited By ~ 1

Author(s):

Eufemia Lella ◽

Ernesto Estrada

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Path Length ◽

Brain Regions ◽

Brain Network ◽

Efficient System ◽

Brain Hemispheres ◽

Disconnection Syndrome ◽

Path Lengths ◽

The communicability distance between pairs of regions in human brain is used as a quantitative proxy for studying Alzheimer’s disease. Using this distance, we obtain the shortest communicability path lengths between different regions of brain networks from patients with Alzheimer’s disease (AD) and healthy cohorts (HC). We show that the shortest communicability path length is significantly better than the shortest topological path length in distinguishing AD patients from HC. Based on this approach, we identify 399 pairs of brain regions for which there are very significant changes in the shortest communicability path length after AD appears. We find that 42% of these regions interconnect both brain hemispheres, 28% connect regions inside the left hemisphere only, and 20% affect vermis connection with brain hemispheres. These findings clearly agree with the disconnection syndrome hypothesis of AD. Finally, we show that in 76.9% of damaged brain regions the shortest communicability path length drops in AD in relation to HC. This counterintuitive finding indicates that AD transforms the brain network into a more efficient system from the perspective of the transmission of the disease, because it drops the circulability of the disease factor around the brain regions in relation to its transmissibility to other regions.

EARLIER DETECTION OF ALZHEIMER’S DISEASE USING IMAGE PROCESSING AND MACHINE LEARNING ALGORITHMS WITH GRAPH THEORY

International Journal of Computer Science and Mobile Computing ◽

10.47760/ijcsmc.2021.v10i08.006 ◽

2021 ◽

Vol 10 (8) ◽

pp. 36-40

Author(s):

D.J. Samatha Naidu ◽

G. Anand Kumar Reddy

Keyword(s):

Machine Learning ◽

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Graph Theory ◽

Thinking Skills ◽

Brain Network ◽

Machine Learning Algorithms ◽

Brain Disorder ◽

Automated Method ◽

Alzheimer’s disease is one of the brain disease which is irreversible, progressive brain disorder that slowly destroys memory and thinking skills and, eventually, the ability to carry out the simplest tasks. There is no cure for Alzheimer’s disease but we prevent it’s by early detection. In existing work, limited with Alzheimer’s are irreversible, effect on daily activities, high memory loss and reducing the size of brain, etc. previous works focused on 2D and 3D formats now we considering 4D images. In proposed work, this work aims to present an automated method that assists in the diagnosis of Alzheimer’s disease supports the monitoring of the progression of the disease. The study of brain network based on resting-state functional Magnetic Resonance Imaging (fMRI) has provided promising results to investigate changes in connectivity among different brain regions because of diseases. Graph theory can efficiently characterize various aspects of the brain network by calculating measures the accuracy of different machine learning methods and different features to classify Cognitively Normal (C.N) individuals from Alzheimer’s Disease (A.D) and to predict longitudinal outcomes in participants with Mild Cognitive Impairment (MCI).

Electromagnetic field in Alzheimer’s disease: a literature review of recent preclinical and clinical studies

Current Alzheimer Research ◽

10.2174/1567205017666201130085853 ◽

2020 ◽

Vol 17 ◽

Author(s):

Reem Habib Mohamad Ali Ahmad ◽

Marc Fakhoury ◽

Nada Lawand

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

In Vivo Studies ◽

Key Factors ◽

Potential Benefits ◽

Preclinical And Clinical Studies ◽

: Alzheimer’s disease (AD) is a neurodegenerative disorder characterized by the progressive loss of neurons leading to cognitive and memory decay. The main signs of AD include the irregular extracellular accumulation of amyloidbeta (Aβ) protein in the brain and the hyper-phosphorylation of tau protein inside neurons. Changes in Aβ expression or aggregation are considered key factors in the pathophysiology of sporadic and early-onset AD and correlate with the cognitive decline seen in patients with AD. Despite decades of research, current approaches in the treatment of AD are only symptomatic in nature and are not effective in slowing or reversing the course of the disease. Encouragingly, recent evidence revealed that exposure to electromagnetic fields (EMF) can delay the development of AD and improve memory. This review paper discusses findings from in vitro and in vivo studies that investigate the link between EMF and AD at the cellular and behavioural level, and highlights the potential benefits of EMF as an innovative approach for the treatment of AD.

Intermittent Hypoxic Conditioning Rescues Cognition and Mitochondrial Bioenergetic Profile in the Triple Transgenic Mouse Model of Alzheimer’s Disease

International Journal of Molecular Sciences ◽

10.3390/ijms22010461 ◽

2021 ◽

Vol 22 (1) ◽

pp. 461

Author(s):

Sónia C. Correia ◽

Nuno J. Machado ◽

Marco G. Alves ◽

Pedro F. Oliveira ◽

Paula I. Moreira

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Mouse Model ◽

Brain Cortex ◽

Special Focus ◽

Mitochondrial Bioenergetics ◽

Related Phenotype ◽

Brain Resilience ◽

The lack of effective disease-modifying therapeutics to tackle Alzheimer’s disease (AD) is unsettling considering the actual prevalence of this devastating neurodegenerative disorder worldwide. Intermittent hypoxic conditioning (IHC) is a powerful non-pharmacological procedure known to enhance brain resilience. In this context, the aim of the present study was to investigate the potential long-term protective impact of IHC against AD-related phenotype, putting a special focus on cognition and mitochondrial bioenergetics and dynamics. For this purpose, six-month-old male triple transgenic AD mice (3×Tg-AD) were submitted to an IHC protocol for two weeks and the behavioral assessment was performed at 8.5 months of age, while the sacrifice of mice occurred at nine months of age and their brains were removed for the remaining analyses. Interestingly, IHC was able to prevent anxiety-like behavior and memory and learning deficits and significantly reduced brain cortical levels of amyloid-β (Aβ) in 3×Tg-AD mice. Concerning brain energy metabolism, IHC caused a significant increase in brain cortical levels of glucose and a robust improvement of the mitochondrial bioenergetic profile in 3×Tg-AD mice, as mirrored by the significant increase in mitochondrial membrane potential (ΔΨm) and respiratory control ratio (RCR). Notably, the improvement of mitochondrial bioenergetics seems to result from an adaptative coordination of the distinct but intertwined aspects of the mitochondrial quality control axis. Particularly, our results indicate that IHC favors mitochondrial fusion and promotes mitochondrial biogenesis and transport and mitophagy in the brain cortex of 3×Tg-AD mice. Lastly, IHC also induced a marked reduction in synaptosomal-associated protein 25 kDa (SNAP-25) levels and a significant increase in both glutamate and GABA levels in the brain cortex of 3×Tg-AD mice, suggesting a remodeling of the synaptic microenvironment. Overall, these results demonstrate the effectiveness of the IHC paradigm in forestalling the AD-related phenotype in the 3×Tg-AD mouse model, offering new insights to AD therapy and forcing a rethink concerning the potential value of non-pharmacological interventions in clinical practice.

Deep-MEG: spatiotemporal CNN features and multiband ensemble classification for predicting the early signs of Alzheimer’s disease with magnetoencephalography

Neural Computing and Applications ◽

10.1007/s00521-021-06105-4 ◽

2021 ◽

Author(s):

Antonio Giovannetti ◽

Gianluca Susi ◽

Paola Casti ◽

Arianna Mencattini ◽

Sandra Pusil ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Brain Regions ◽

Ensemble Classification ◽

The Novel ◽

Ensemble Classifiers ◽

Deep Convolutional Neural Networks ◽

Early Signs ◽

Data Representations ◽

AbstractIn this paper, we present the novel Deep-MEG approach in which image-based representations of magnetoencephalography (MEG) data are combined with ensemble classifiers based on deep convolutional neural networks. For the scope of predicting the early signs of Alzheimer’s disease (AD), functional connectivity (FC) measures between the brain bio-magnetic signals originated from spatially separated brain regions are used as MEG data representations for the analysis. After stacking the FC indicators relative to different frequency bands into multiple images, a deep transfer learning model is used to extract different sets of deep features and to derive improved classification ensembles. The proposed Deep-MEG architectures were tested on a set of resting-state MEG recordings and their corresponding magnetic resonance imaging scans, from a longitudinal study involving 87 subjects. Accuracy values of 89% and 87% were obtained, respectively, for the early prediction of AD conversion in a sample of 54 mild cognitive impairment subjects and in a sample of 87 subjects, including 33 healthy controls. These results indicate that the proposed Deep-MEG approach is a powerful tool for detecting early alterations in the spectral–temporal connectivity profiles and in their spatial relationships.